---------------------------------------------- ####### ######## ######## ########### ### ### ## ### ## # ### # Interpersonal Computing and ### ### ## ### ## ### Technology: ### ### ## ### ### An Electronic Journal for ### ######## ### ### the 21st Century ### ### ### ### ### ### ### ## ### ISSN: 1064-4326 ### ### ### ## ### April, 1993 ####### ### ######## ### Volume 1, Number 2 --------------------------------------------------------------------------- Published by the Center for Teaching and Technology, Academic Computer Center, Georgetown University, Washington, DC Additional support provided by the Center for Academic Computing, The Pennsylvania State University, University Park, PA 16802 This article is archived as KUEHN IPCTV1N2 on LISTSERV@GUVM (LISTSERV@GUVM.GEORGETOWN.EDU) --------------------------------------------------------------------------- COMMUNICATION INNOVATION ON A BBS: A CONTENT ANALYSIS Scott A. Kuehn Clarion University of Pennsylvania ABSTRACT The nature of innovative communication behaviors on a computer- mediated bulletin board system was investigated using content analysis. The play theory of communication was used to analyze messages for content that demonstrated special manipulations of computer symbols to expess individuality and originality. Three categories of message types achieved intercoder reliability: textual manipulations signifying jokes, irony, witticism, and sarcasm, conventional text emphasis symbols used to convey conversational tone and nonverbal communication, and unconventional symbolic representations of nonverbal communication. Messages revealed that BBS users manipulated text symbols, shifting from simple display of text to textual displays of originality and personal innovation. These findings contradict a "cues-filtered-out" approach to understanding CMC and demonstrate that innovative CMC which adapts to social contexts aids in defining self to self and self to others. INTRODUCTION Electronic mail is often considered one innovation of the computer revolution. With E-mail, computer users can send messages through networks, and read them on their computers. While the ability to do this on a mainframe computer has been around since the 1960s, micro-computer users have introduced innovations with broad implications for communicators. Micro- computers provided an inexpensive means to set up and operate a bulletin board service (BBS) within one's own home. With the aid of a modem and telephone line, a "personal" computer can become a mediated "interpersonal" computer, providing the hub for a groundswell of communication activity. The purpose of this research was to examine BBS messaging to describe medium- specific communication behavior for evidence of innovation and adaptation to the CMC context. The BBSs has become popular forums of interaction. Computer owners can connect with a modem and a phone line to another computer to leave messages or interact "live," online, to another user on another phone line. Today it is estimated that there are over five thousand BBSs operating around the country, with more than 1500 installed in corporations for dealing with customer support (Rapaport, 1991). The ability of a BBS to support new users invites initiation to computer-mediated communication and many new computer systems come packaged with modems and instructions on how to connect with BBSs. In addition to privately run systems, some BBSs are set up by schools (Alifrangis, 1988), and private organizations use BBSs to support the activities of their members. The communication opportunities provided by BBSs fall into two categories: 1) "live" real time interaction with a BBS host or fellow user, or 2) "asynchronous" e-mail type exchange of messages. "Real time" interaction is limited by the number of phone lines (and thus the number of simultaneous users of a system). Callers on one line can talk to callers on another line, or to the BBS operator. E-mail is the most common form of BBS communication. A further innovation is the relay network. These networks establish connections between BBSs so that mail messages may be forwarded to other areas of the country, to users on distant BBSs. Fidonet was the first such service, followed by others (Rapaport, 1991). With the exception in product support and corporate systems, most BBS users communicate in a non-task oriented, active environment. This is in contrast to work environments where computer-mediated communication, often on mainframe systems or local area networks, is connected with job tasks. The appeal of BBS communication is thus hard to explain in theoretical models that place heavy emphasis on contextual limitations of CMC in organizations or groups. Walthers (1992) argues that many of these group and organizational CMC studies fail to consider the limitations of the work related context. In a review of research issues in computer-mediated communication, Rice (1988) points to the role of "stakeholders" in guiding inquiry into the use of CMC systems. But, in the BBS CMC context, stakeholders have nothing more at stake than the desire to interact. Walther (1992) presents a compelling argument that many studies of CMC in organizational contexts have failed to characterize non-task oriented situations. Often, according to Walther, the emphasis of these studies is on reductive features of CMC, following a "cues-filtered-out approach" (1992, p. 60; see Hiltz, Johnson, & Turoff, 1986; Connolly, Jessup, & Valacich, 1990; and Rice and Love, 1987) BBS CMC, however, presents a more socially oriented context than organizational CMC (Rice, 1987). Rice and Love (1987) found a significant amount of socio-emotional messaging in BBS contexts, more than they had found in their review of previous studies. In addition, socially oriented relationships can develop on-line. Smelter's (1992) content analysis of BBS messages found complexity, messages averaged five sentences, with approximately 80% of the words used only once (low repetition) and average messages had a FOG Index of 8.7, indicating a ninth grade reading level (p. 53). Shamp (1991) found evidence that the more personal information exchanged on a BBS, the less likely relational partners would mechanomorphize fellow interactants (perceive them as computer-like). In the non-task oriented BBS situation, a focus on the role of the user as communication innovator can shed light on a number of communication behaviors that are common only in CMC. Walther (1992) posits that a central variable in understanding CMC is motivation to communicate: CMC users, just as communicators in any context, should desire to transact personal, rewarding, complex relationships and that they will communicate to do so (p. 68). The research task at hand is to discover the means by which communicators develop relational skills on-line through an adaptive set of behaviors fitting the characteristics of the CMC medium. In organizational studies examples of such behavior include placing "signatures" at the end of messages, which were found to indicate hierarchical status (Sherblom, 1988). Similarly, "carbon-copying" messages to superiors was found to be a compliance gaining strategy (Phillips and Eisenberg, 1989). Typing-related behaviors, unique to CMC, have also been noted, such as "relational icons" or "emoticons" (combinations of punctuation marks that form smiling or frowning faces) (Asteroff, 1987; Sherblom, 1988). Carey (1980) noted the function of misspelling, capitalization and grammatical markers as emphasis cues in CMC, along with lexical surrogates (such as gasp, or ohh) and repetious punctuation marks (!!!!!). PLAY THEORY Why does a computer user choose to extend his/her communication experience through computer-mediated communication? Perhaps, as Katz and Blumler (1974) argue for other media, the BBS provides uses and gratifications to the communicator. Play theory provides a framework that can illuminate uses and gratifications in a BBS environment. Introduced in 1967 by William Stephenson, play theory focuses on the media user's relationship with the medium. Stephenson argued that "play" involved the ability to select among a number of products, ideas, issues, and entertainment options from the media. According to Stephenson, media content is used to engage in "convergent selectivity" -- making choices about media allows us to express ourselves. Our patterns of choices provides a feeling of individuality that we use to subsequently understand ourselves. This is Stephenson's concept of "communication play," a form of self-expression that takes advantage of the various media to promote a feeling of individuality. While Stephenson found communication play a uniquely personal experience, Davis and Baran (1981) extended the play concept into interaction endeavors. Using Goffman's (1974) definitions of play and "keyings," Davis and Baran argued that there are two types of communication play: participatory play and elite dominated communication play. Participatory play takes place when communicators are free to alter interaction to achieve their goals. On the other hand, elite dominated communication play is a passive activity. Here play exists only through making choices of what to use from existing media content. Participatory play provides a center of interest when considering the example of BBS computer-mediated communication. The medium of play becomes the computer-teleconnection, the content of play is determined by the participants. Thus, two types of play choices can take place during computer-mediated communication: 1) the decision to use the telelink context and, 2) unique features of the link that provide content cues of communication play. We can see in this perspective that much organization based, task-centered CMC would be elite-dominated communication play, whereas the populous BBS network would provide a freer system of expression. The decision to use a BBS could thus be an initial expression of a desire to initiate a form of communication play. Goffman's (1974) explanation of interaction play can provide a helpful means to analyze the content cues of communication play. Goffman believed that "play" was a situation grounded on a transformation of a serious situation. Thus "play" fighting takes place with wooden swords with mimicked moves. The transformation from serious to "play" situation is rule governed and takes place as long as the participants observe the rules. Within the rules of the play situation, cues are provided that establish and maintain the transformation to play. Goffman regarded these cues as "keyings," (similar to the music metaphor) that signaled a special form of interaction based on a serious type. Among the types of behavior that Goffman considered play are jokes, dramas, sports contests, games, simulations, practices, and role-playing. An analysis of the content of play communication could begin with the cues of play activity, referred to as "keyings". The goal here is to operationalize content cues that demonstrate shifts from simple display of text to attempts to manipulate characteristics of the electronic text medium to display originality and personal innovation. These can range from textual jokes and irony to manipulation of keyboard symbols to emulate conversational tone and nonverbal communication. Content analysis can be employed to explore three types of keyings: 1) textual manipulations signifying jokes, sarcasm, irony, and witticism. 2) conventional text emphasis cues to signal nonverbal communication information, such as underlining, capitalization, slang, exclamation points, etc. 3) unconventional symbolic representations of nonverbal communication, such as a "smiley" (:-) and other constructions. An analysis of BBS keying behavior provides insight into the adaptations made in the CMC medium to promote individuality and personal expression, prime "play" goals according to Stephenson (1967). METHOD Content analysis was used in this study to describe the types of keying behavior BBS communicators used for communication play. A BBS was selected at random, a group of messages were selected at random on the selected BBS, and coders examined three analysis units to reveal keyings in message content. A random number generator provided a list of numbers to use in the selection of a sample BBS from a sampling frame of BBSs. The sampling frame was a list of IBM PC compatible BBSs compiled and periodically updated by a BBS in the eastern U.S. This list is widely disseminated across the U.S. on BBSs that use IBM compatible hardware. Criteria for being on this list include having one or more modems that operate at 2400 baud or faster, and at least 60 megabytes of hard disk storage. Over 500 BBSs are on this national list, and are placed by registering with the BBS that produces the list. The BBSs on the list were numbered, and the selected case was a BBS in New Jersey that specialized in scientific topics. The BBS was contacted, and the message base consulted. Most BBSs have numbered messages, posted publicly, and allow users to select and download these messages. This was true for the sample BBS. Messages examined for this study were posted as public. BBSs usually post a notice to users that messaging is never private (as a system operator will examine them) to relinquish any liability for public access to private communication. The sample BBS posted such a message and made joining the BBS as a user contingent upon acknowledgment of the public nature of messages. A second computer generated random number provided a message number starting point. Fifty messages following this randomly selected message were downloaded for analysis. Three units of analysis used in this study examined the keyings shifting into play communication on text messages. These included: 1) Textual manipulations signifying jokes, irony, witticism, and sarcasm. 2) Conventional text emphasis symbols that were used to convey conversational tone and nonverbal communication. Included were underlinings, exclamations, misspellings for slang purposes, capitalizations for emphasis. 3) Unconventional symbolic representations of nonverbal communication. These could be unique combinations of keyboard symbols to give readers a view of the author's facial expressions, verbalized pauses "," and possibly gestures. CODING Two messages from the sample of 50 were incomplete, and one message was taken from the sample to use for coder instruction. Forty-seven complete BBS messages were printed on paper and given to ten coders. Before coding, coders were shown the sample message and were instructed on how to recognize each of the three units of analysis. They were given color coded pens and instructed to circle each instance of each of the units of analysis as they encountered it, reading the material at a normal pace. Coders were given forty minutes to code the messages. Two coders who finished before 30 minutes were deleted from analysis, as was another coder who did not finish coding within forty minutes. Seven coders were included in the final analysis. RESULTS The coding sheets were tabulated by coder for each of the units of analysis. A chi square contingency table was constructed that matched the three unit categories with the seven coders. There was no significant difference between the coders in their codings of the three types of analysis units (ChiSq = 18.11; D.F. = 12; P = .117). Subsequent analysis employed a chi square test of goodness of fit to test the reliability of each coding unit. There were no significant differences between the coders on any of the three units of analysis entered separately: jokes, irony, witticism (ChiSq = 5.62; D.F. = 6; P = .468), conventional manipulation of text (ChiSq = 8.11; D.F. = 6; P = .232), and unconventional manipulation of symbols (ChiSq = 10.29; D.F. = 6; P = .115). Thus, there was inter-coder reliability. Among the coders there was an average of 45 observed instances of jokes, irony, and witticisms, 45 observed instances of conventional text manipulations to indicate nonverbal communication, and 34 observed instances of unconventional symbolic constructions to indicate nonverbal communication. A total of 3240 words were in the message base. Many of the jokes and witticisms in the messages were in the form of sign-offs, phrases placed at the end of messages. Instances included: Sex in a Volkswagen = Farfergnookie. I larned eferythang I knou in hai skool. Support medical examiners - die strangely. We tortured the data until it confessed. Does the name "Pavlov" ring a bell? The Business of America is Beancounting. Use DEVICE=EXXON to screw up your environment We're OVERSTOCKED With Strong Backs. LIFE Begins When You Start Minding Your Own Business. Others included references to issues and general topics of interest: Sometimes I get up from my desk, and by the time I go out the door, I forgot what I got up for! _Uh, what was my name again?_ How about "The Orka Winfreed Show"? Inquiring minds want to know! Being the good little telecommunicator that I am He's quite opinionated (unlike some folks we know ... Yeah, Sen. Phil Gramm and your lovely, (no "pull in Federal appointments, here") wife Wendy - the sleazy money bunney. Underlinings, capitalizations, and slang misspellings characterized conventional manipulations of text. Since there is no way to provide regular underlining in electronic ascii text, users would underline before and after a phrase for emphasis: So I knew there had to be _something_ wrong! You learn to _make_ time. that infection can take place _only_ via infected floppy disks. Capitalizations played a similar emphasis role to underlining, if yet to signify even more emphasis: The other downside to the narrow-spectrum approach is that uninformed individuals will consider their environments "safe" because THE virus has been removed. If this misconception results in MORE FREQUENT unfiltered disk and file-swapping activities, the probability of .COM and .EXE, not to mention boot- sector viruses, will increase exponentially. Americans are only willing to buy virus-free software - do you know how much that'll COST?! Slang misspellings seemed to get the most attention, and were seemingly the most playful of the constructions: Right now, I'm considering a move to a better apartment complex - you can rest assured that it won't be one within the "Citah" Limits. Whatsamatta, you a friggin' GROWNUP or sumpthin'?! OOooh hot damn! You got that right. 'S too bad that most folks depend on The Media for the basic news information I caught a bit of a news item on TeeVee the other nite, F'r example, did you know that there are lotsa viruses which won't attach Holy Cheeses! How did these guys let that happen?! Verrrry interesting. Dunno. It's certainly not "safe computing" these days Mebbe that's because we don't have alot of Seems to me that you'd prolly do better Unconventional manipulation of symbols took two forms. One type mimicked facial expressions, usually to signify sarcasm: I didn't know that a System 36 used Ms-Dos 3.3 :-) another type of construction used letters within brackets to signify nonverbal behavior, like , which signified a grin, or and which signified vocalized pauses: A guy I work with was telling me that he was telling his son that he (my buddy) was losing his short-term memory. The son (a wiseass, to be sure) said something like, "Hey, dad! Look at the benefits: You'll always be meeting new people, and you can even hide your own Easter eggs!" Could it be that it's another media- blown non-event? DISCUSSION This research demonstrated that users who communicate for enjoyment with computers manipulate text to enhance their communication. Play theory demonstrates that much of the gratification of telecommunication with computers is through manipulating text symbols to promote individuality. The "fun" of this brand of CMC appears to be in figuring out context bound ways of self expression. It is interesting to note the subtle differences in tone that can be achieved with these textual manipulations. Style differences can be discerned between communicators. Differing brands of sarcasm, wit, and humor are only one of these style distinctions, although the most readily apparent. Other emphasis keyings provide a view into emotional commitment, as when reasons are given capitalizations or underlinings, or when unique nonverbal representations cue a reader that a point is not to be taken seriously. These adaptations add to the richness of the communication transaction and provides evidence, following Walther's (1992) critique, that the "cues-filtered-out" approach is limited when explaining CMC in non-task oriented contexts. Play theory alerts us to the fact that much of the gratification of communicating is using communication behavior as a way of defining "self." It is apparent when examining BBS communication that much of it emulates telephone conversation, with text adaptations to provide richer communication behavior within the CMC context. The question then is, why don't users simply call each other to communicate, rather than using text? Part of the reason may be from the asynchronous nature of the transaction -- a person can leave a message for another who will see it when they log on. But the gratification seems to go beyond the choice of the medium. Recall that while each of these messages were addressed to a single user, they were marked "public." It could be that the gratification of defining self as a computer communicator, as well as displaying one's talent as a computer communicator, is a predominant gratification. This study is somewhat limited in breadth. The symbols revealed in this sample may not be indicative of typical manipulations, but the fact that they exist provides strong evidence that this behavior occurs more often. A larger, more demographically diverse sample may reveal other types of adaptive text manipulations. In addition to expanding the sample under examination, further study should be conducted to test the effect of time on communication innovation in CMC, a variable Walther (1992) calls chronometry. It is likely that most CMC behavioral adaptations are learned from the social context, and patterns of keyings that develop over time could reveal the process by which new users become acculturated on BBSs. The study could proceed with examination of messages marked over a period of time, with emphasis on development of keying patterns revealing innovative communicator style. This study demonstrated that CMC behavior is not simply a matter of sharing text, but of adopting ways of manipulating symbols to enrich electronically posted text. Further study of CMC textual manipulation should be grounded in theory that demonstrates the richness, not paucity, of communication cues in the CMC context. References Alifrangis, C. (1988, April). A descriptive analysis of student messages on a school district's electronic bulletin board. Paper presented at the meeting of the Eastern Communication Association, Baltimore, Maryland. Asteroff, J. F. (1987). Paralanguage in electronic mail: A case study. Unpublished doctoral dissertation, Columbia University, New York. Carey, J. (1980). Paralanguage in computer mediated communication. In N. K. Sondheimer (Ed.), The 18th annual meeting of the association for computational linguistics and parasession on topic in interactive discourse: Proceeding of the conference (pp. 67-69). Philadelphia: University of Pennsylvania. Connolly, T., Jessup, L. M., & Valacich, J. S. (1990). Effects of anonymity and evaluative tone on idea generation in computer- mediated groups. Management Science, 36, 97-120. Goffman, E. (1974). Frame Analysis. New York: Harper & Row, Publishers. Hiltz, S. R., Johnson, K., & Turoff, M. (1986). Experiments in group decision making: Communication process and outcome in face- to-face versus computerized conferences. Human Communication Research, 13, 225-252. Katz, E; Blumler, J. G.; and Gurevitch, M. (1974). Utilization of mass communication by the individual. In The uses of mass communication, ed. J. G. Blumler and E. Katz, pp. 19-34. Beverly Hills, CA: Sage. Phillips, S. R., & Eisenberg, E. M. (1989, May). The persuasive use of electronic mail in organizations. Paper presented at the annual meeting of the International Communication Association, San Francisco. Rapaport, M. (1991). Computer Mediated Communications: Bulletin Boards, Computer Conferencing, Electronic Mail, Information Retrieval. New York: John Wiley & Sons. Rice, R. E. (1987). Computer--Mediated Communication Systems and Organizational Innovation. Journal of Communication, 37:64-94. Rice, R. E. (1988). Issues and Concepts in Research on Computer- -Mediated Communication Systems. In Communication Yearbook, Vol. 12, ed. J. Anderson, pp. 436-476. Beverly Hills, CA: Sage. Rice, R. E., & Love, G. (1987). Electronic emotion: Socioemotional content in a computer-mediated network. Communication Research, 14, 85-108. Shamp, S. A. (1991). Mechanomorphism in perception of computer communication partners. Computers in Human Behavior, 7, 147-161. Sherblom, J. (1988). Direction, function, and signature in electronic mail. Journal of Business Communication, 25, 39-54. Smeltzer, D. K. (1992). Computer-mediated communication: An analysis of the relationship of message structure and message intent. Educational Technology, June, 1992, 51-54. Stephenson, W. (1967). The play theory of mass communication. Chicago: University of Chicago Press. Walther, J. B. (1992). Interpersonal effects in computer- mediated interaction: A relational perspective. Communication Research, 19, 52-90. ------------------------------------------------------- BIOGRAPHICAL NOTE: Scott Kuehn (Ph.D. 1987, The Pennsylvania State University) is an assistant professor in the Department of Communication at Clarion University of Pennsylvania. He is also the director of the Center for the Study of Computer Communication. Inquiries and comments on this article can be sent to SKUEHN@VAXA.CLARION.EDU. --------------------------------------------------------------------------- Interpersonal Computing and Technology: An Electronic Journal for the 21st Century Copyright 1993 Georgetown University. Copyright of individual articles in this publication is retained by the individual authors. Copyright of the compilation as a whole is held by Georgetown University. It is asked that any republication of this article state that the article was first published in IPCT-J. Contributions to IPCT-J can be submitted by electronic mail in APA style to: Gerald Phillips, Editor IPCT-J GMP3@PSUVM.PSU.EDU